Recommendations for search queries

Abstract

A system and method of providing recommendations for search queries are disclosed. A search query is received. Historical data in a historical database is accessed using the search query. The historical data is related to inactive e-commerce items. Recommendations are generated for the search query based on the accessed historical data. The recommendations can comprise item recommendations, category recommendations, and query recommendations. The query recommendations can comprise suggested queries and a preview of search results for each of the suggested queries. The steps of accessing the historical data and generating the one or more recommendations can be performed in response to a determination that the number of results for the search query is below a predetermined threshold.

Description

RELATED APPLICATIONS[0001]This application claims priority to U.S. Provisional Application No. 61 / 538,403, filed on Sep. 23, 2011, and entitled, “ITEM RECOMMENDATIONS FOR LOW AND ZERO RESULT QUERIES,” which is hereby incorporated by reference in its entirety as if set forth herein.TECHNICAL FIELD[0002]The present application relates generally to the technical field of search query processing, and, in various embodiments, to systems and methods of generating recommendations for search queries.BACKGROUND[0003]When trying to find an item on an e-commerce site, users often type a query into the site's search engine, which attempts to match that query against all inventory that is active on the site. Active inventory is currently available, whereas inactive inventory is not available, as it has expired or has already been sold. Although matches for the query are often found and shown to the user, there are also many situations where the site is unable to match the query to any items or o...

AI Technical Summary

Benefits of technology

The present invention relates to a system and method for generating recommendations for search queries on e-commerce websites. The technical effect of this invention is to improve the user experience by providing relevant recommendations for low- and zero-result queries, which are often the result of a user's search engine query. The system accesses historical data related to inactive e-commerce items and generates recommendations based on that data to suggest relevant items or queries for the user. This helps to better match the user's intent and provide a better search experience.

Problems solved by technology

Active inventory is currently available, whereas inactive inventory is not available, as it has expired or has already been sold.

Although matches for the query are often found and shown to the user, there are also many situations where the site is unable to match the query to any items or only able to match the query to a few items.

Specifically, where a particular seller has authored and / or published a large number of listings, the management of such listings may present a challenge.

In one example, a user enters “turquoise peace sign and cross” as her search query, but it is not initially matched to any available inventory.

For many such queries, the search engine 350 cannot find many matching items because the queries are over-constrained, specialized, or too verbose.

In many cases, generalization can lead to loss of information or precision.

Additionally, searching against longer contexts like descriptions instead of shorter titles can also lead to a loss of precision.

Images